At some point, in Homework 3, you will encounter the integral
For those of you who took complex analysis, this should be easy to do, using the methods of contour integration.
If you haven’t taken complex analysis, the answer is
To see how the integral is done, read on…
First, let’s rewrite
The displaces the location of the poles of the integrand off the real axis. The are located at and at . The contour can be closed either in the upper half-plane or in the lower half-plane. Closing in the upper half-plane, we pick up the pole at :
and we get . If we closed the contour in the lower half-plane, we would pick up the pole at . But, that contour is in the clockwise (instead of counter-clockwise) direction, the two minus-signs cancel, and we get the same result for the integral.
I think you’re asking about how to relate Cartesian and spherical coordinates.
Our formula for Lorentz boosts was written in Cartesian coordinates. What we’re interested in is the relation between the corresponding systems of spherical coordinates.
The special case, where all the Jordan blocks are , is the case where is diagonalizable. When is diagonalizable, the matrix has columns, where the column is the eigenvector of , corresponding to the eigenvalue :
B v_k = \lambda_k v_k
or, equivalently,
(B-\lambda_k\mathbb{1}) v_k=0
Of course, this is a little bit ambiguous, as we can always multiply each by a non-zero constant. That ambiguity drops out of (1).
When a Jordan block has size , we need, not one, but vectors to make up the corresponding columns of . The first column is, again, given by the eigenvector
(B-\lambda_k\mathbb{1}) v_k=0
The next column is given by a vector, , which satisfies
(B-\lambda_k\mathbb{1})w_{k,2} = v_k
The column after that is given by , which satisfies
(B-\lambda_k\mathbb{1})w_{k,3} = w_{k,2}
and so on. Again, each of the is ambiguous by the addition of a multiple of , but this drops out of (1).
In conversation at Office Hours, it came up that not every integral can be written in terms of elementary functions.
In particular, there’s a function called the “Error Function,”
Erf(s) = \frac{2}{\sqrt{\pi}} \int_0^s e^{-t^2} d t
Of course, we have that is a monotonic function, with and (more nontrivially) .
To prove the latter, note that
\begin{split}
{\left(\int_0^\infty e^{-u^2} du\right)}^2 &= \int_0^\infty\int_0^\infty e^{-(u^2+v^2)} d u d v\\
&= \int_0^\infty e^{-r^2} r d r \int_0^{\pi/2} d \theta\\
&=\frac{\pi}{4} \int_0^\infty e^{-x} d x\\
&=\frac{\pi}{4}
\end{split}
Here is a table summarizing how various thermodynamic quantities change, when an ideal gas undergoes various processes.
Process
- curve
isothermal ()
adiabatic ()
iso-volume ()
isobaric ()
The basic equations are
\begin{aligned}
&\text{Ideal Gas Law:}\quad&p V &= N k T = \frac{1}{\alpha} U\\
&\text{First Law:}\quad&\Delta U &= Q - W\\
&\text{Entropy:}\quad&Q &= T \Delta S\\
&\text{Work:}\quad&W &= p \Delta V \\
\end{aligned}
The cards don’t fall, provided the total force and total torque on each card vanishes. By symmetry, we can examine the forces and torque on just one of the cards.
Let’s take the point of contact with the floor as the axis about-which to evaluate the torque. Then only two of the four forces (gravity and the force due to the other card, ) contribute to the torque. If we call the length of the card, , the displacement at which gravity acts is , and the perpendicular component of the gravitational force is . The force from the other card acts at a displacement , and the perpendicular component is . So the total torque is
(1)\tau = (N_c \cos \theta -\tfrac{1}{2} m g \sin\theta ) L
The components of the total force are
(2)\begin{split}
F_x &= F_f - N_c\\
F_y &= F_N - m g
\end{split}
We also must have
(3)|F_f|\leq \mu_s F_N
Combining (3) with (2), we have . Making the torque vanish in (1) requires
N_c = \tfrac{1}{2} m g \tan\theta
So we need
\tfrac{1}{2} m g \tan\theta \leq \mu_s m g
or
\tan\theta \leq 2 \mu_s
If we try an angle larger than that, the cards will slide.
In class, we considered only the case where one of the masses is much greater than the other, , and so is more-or-less stationary. What happens when that isn’t the case?
Let’s consider the extreme opposite situtation, namely two equal masses, , in a circular orbit about the midpoint between them.
The distance between them is , so the force on the first one is
F_1 = \frac{G m^2}{{(2R)}^2}
(and, similarly, for ). So, for a circular orbit, we have
\begin{split}
U_{\text{grav}}(R_E+y) &= -\frac{G M_E}{R_E}\left(1 -{\left(\tfrac{y}{R_E}\right)} + {\left(\tfrac{y}{R_E}\right)}^2 -\dots\right)\\
&= - \frac{G M_E}{R_E} + m g y - m g \tfrac{y^2}{R_E} + \dots
\end{split}
The first term is an irrelevant constant. The second is the gravitational potential energy we wrote down before, and the third term is the first correction, due to the fact that the gravitational force, due to the Earth, falls off with distance (rather than being constant).
I have gotten dozens of emails, asking questions about the Practice Exam. I’ll give the answers to those questions here.
No, the Practice Exam does not count for a grade.
Yes, it has a due-date (4 pm Sep 26, in the case of the 1st one).
Yes, the solutions are available, as soon as the due-date has passed (just as they are for the homeworks).
No, I’m not going to change the due-date. The class seems to be divided between those who have taken the practice exam — and want to see the solutions — and those who haven’t taken it. The due-date was chosen so as to give everyone the maximum opportunity to do the practice exam, while still giving folks the opportunity to also study the solutions.
Update:
As it turns out, Quest had a bug (fixed now) which prevented students, who had not taken the practice exam before the due-date, from viewing the solutions. The bug is fixed now – alas, too late for the the 1st exam.
Because the Quest website seems to be having major problems, I have delayed the due-date of 2nd homework by 24 hours. I hope that they have things sorted out by then. Sorry for the inconvenience …
I upgraded the software, on which this forum runs (to Rails 3.1.0 if you want to know the geeky details). The process didn’t go as smoothly as I would have liked, and service was pretty disrupted for most of Friday.
Should be back to normal now. But leave a comment here, if something’s still broken for you.